Search

CN-122025133-A - Individual health early warning method and system integrating physiological monitoring and behavior questionnaires

CN122025133ACN 122025133 ACN122025133 ACN 122025133ACN-122025133-A

Abstract

The invention relates to the technical field of health detection and early warning. The individual health early warning method comprises the steps of collecting user physiological, behavior and environment data, constructing an individual health knowledge map fusing medical knowledge, establishing a digital twin model fusing mechanism and data drive based on the map, carrying out dynamic risk assessment and tracing, simulating deduction in the digital twin according to the result to generate a personalized intervention plan, and forming closed loop optimization. The system comprises an edge perception layer, a cloud intelligent central layer, an application service layer and an ecological interface layer. The invention realizes the whole-flow health management from multi-source data fusion, intelligent risk assessment to personalized intervention closed loop, and improves the accuracy, foresight and effectiveness of health early warning.

Inventors

  • XU LIANWEN

Assignees

  • 北京普济华堂健康科技有限公司

Dates

Publication Date
20260512
Application Date
20260130

Claims (9)

  1. 1. An individual health early warning method integrating physiological monitoring and behavior questionnaires is characterized by comprising the following steps: S1, personalized configuration and multi-source data acquisition, namely, based on a user individual file, self-adapting a data acquisition mode, acquiring physiological time sequence data through household monitoring equipment, and acquiring behavior data and environment data through a dynamic interactive questionnaire; S2, data fusion and personal health knowledge graph construction, namely preprocessing and semantically analyzing the acquired physiological time sequence data and behavior data, and constructing and dynamically updating the personal health knowledge graph based on the preprocessed physiological time sequence data, behavior data, environment data and a medical knowledge base, wherein the personal health knowledge graph takes a user entity as a core, associates physiological index nodes, behavior event nodes and environment factor nodes, and defines time sequence, cause and effect and statistical association relations among the nodes; S3, risk assessment based on digital twinning and personal health knowledge graph, namely establishing a personal health digital twinning model of a user based on the personal health knowledge graph, carrying out multi-dimensional and hierarchical dynamic health risk assessment on the user by utilizing the personal health digital twinning model and a risk assessment algorithm, and outputting a risk assessment result and a risk tracing path; and S4, personalized intervention and closed loop verification, namely generating and pushing a personalized health intervention plan according to the risk assessment result and the risk tracing path, tracking feedback data after the user executes the intervention plan, quantitatively assessing the intervention effect, and dynamically optimizing the intervention plan and the personal health knowledge graph based on the assessment result.
  2. 2. The method for early warning of individual health by combining physiological monitoring and behavioral questionnaires according to claim 1, wherein in step S1: The household monitoring equipment comprises a sphygmomanometer, a glucometer and intelligent wearing equipment which support one-key operation and voice guidance and are suitable for ageing design, and also comprises a non-contact sensor for monitoring respiratory frequency, body movement and bed leaving behaviors; The implementation mode of the dynamic interaction questionnaire comprises scene embedded triggering, voice interaction, image auxiliary identification and a hook-up type simplified option, wherein the image auxiliary identification comprises photographing identification of food and medicine packaging, and the dynamic interaction questionnaire supports self-adaptive optimization of problem content and presentation mode according to user response history and current scene.
  3. 3. The method for early warning of individual health by fusing physiological monitoring and behavioral questionnaires according to claim 1, wherein in step S2, constructing an individual health knowledge graph specifically includes: s21, performing entity extraction and emotion analysis on unstructured text information in behavior data to form a structured behavior event sequence; S22, carrying out association mapping on the physiological time sequence data, the structured behavior event sequence and the environment data and the disease, symptom and risk factor relationship from a medical knowledge base; s23, processing the data after the association mapping by using a causal discovery algorithm to infer potential causal relations among nodes, wherein the causal relations are used as the basis for constructing and optimizing the association edges in the personal health knowledge graph.
  4. 4. The method for early warning of individual health by combining physiological monitoring and behavioral questionnaires according to claim 1, wherein in step S3: The personal health digital twin model adopts a mixed architecture combining a mechanism model and a data driving model and is used for simulating the response of the physiological state of a user to behavior and environment input; the risk assessment algorithm comprises a basic layer algorithm for single-dimensional anomaly monitoring, a comprehensive layer algorithm for multi-factor correlation analysis and risk conduction network analysis, and a dynamic layer algorithm for dynamically adjusting risk weights based on historical trends and real-time data; The risk tracing path is obtained by reversely traversing the associated edges from the high-risk index nodes on the personal health knowledge graph and identifying a key influence factor chain.
  5. 5. The method for early warning of individual health by fusing physiological monitoring and behavioral questionnaires according to claim 4, wherein the step S3 further comprises: And the human-computer collaborative knowledge base is introduced to calibrate and correct the threshold value and the result of the risk assessment, integrates clinical guideline standards, can receive feedback information of an expert on a system case, and uses an active learning mechanism to optimize the risk assessment algorithm and the personal health knowledge graph.
  6. 6. The method for early warning of individual health by fusing physiological monitoring and behavioral questionnaires according to claim 1, wherein in step S4, generating a personalized health intervention plan specifically includes: S41, based on the risk assessment result and the risk tracing path, performing simulation deduction of various intervention measures in the personal health digital twin model, and predicting the effect and potential risk of each measure; S42, selecting an optimal intervention measure set according to a simulation deduction result and forming a personalized health intervention plan containing specific execution tasks and targets; S43, explaining the intervention plan, decomposing daily tasks and providing execution guidance and confirmation to a user in a natural language interaction mode through the interactive health assistant.
  7. 7. The method for early warning of individual health by fusing physiological monitoring and behavioral questionnaires according to claim 6, wherein in step S4, the quantitative evaluation of the intervention effect is specifically: The method comprises the steps of establishing an intervention and response effect model, quantifying the effectiveness of an intervention plan by comparing the state change of relevant nodes and the change of risk assessment of a user in the personal health knowledge graph before and after the execution of the intervention plan, and deciding whether to maintain, strengthen and adjust the intervention plan or trigger higher-level early warning according to the quantified result.
  8. 8. An individual health warning system incorporating physiological monitoring and behavioral questionnaires, for implementing the method of any one of claims 1-7, the system comprising: the edge perception layer comprises physiological monitoring equipment, a behavior data acquisition terminal and an edge computing gateway, wherein the physiological monitoring equipment, the behavior data acquisition terminal and the edge computing gateway are deployed on a user side, and the edge computing gateway is used for equipment access management, local data preprocessing, lightweight real-time anomaly detection and data encryption uploading; The intelligent cloud center layer comprises a data federation and storage module, an intelligent analysis engine, a man-machine collaborative knowledge base, a data analysis module and an intervention scheme generation module, wherein the cloud intelligent center layer is used for receiving and storing the encrypted characteristic data of the edge perception layer and supporting distributed learning under privacy protection; The system comprises an application service layer, a health assistant agent and a user terminal application, wherein the application service layer comprises a user terminal application, a monitoring end application and the health assistant agent, the user terminal application is provided with a context awareness type self-adaptive interface, and can adjust an interaction mode according to a user state; and the ecological interface layer is used for interfacing with a third-party health service system and supporting service reservation and data sharing based on early warning results or intervention plans.
  9. 9. The individual health early warning system integrating physiological detection and behavior questionnaires according to claim 8, wherein the knowledge graph management unit is used for maintaining and updating an individual health knowledge graph with a user entity as a core, and the digital twin modeling unit is used for constructing and running an individual health digital twin model of a user for risk simulation and intervention deduction through the knowledge graph management unit and user history data.

Description

Individual health early warning method and system integrating physiological monitoring and behavior questionnaires The invention relates to the technical field of health monitoring, in particular to an individual health early warning method and system integrating physiological monitoring and behavior questionnaires. Background Along with the acceleration of population aging process and popularization of health management concepts, individual health early warning has become an important research direction in the field of public health, and the core aim is to identify potential health risks in advance through the collection and analysis of individual health related data so as to provide support for accurate intervention. At present, the existing individual health early warning technology mainly develops around two major directions of physiological index monitoring and health information investigation, wherein the physiological index monitoring is mainly based on household monitoring equipment (such as a sphygmomanometer, a glucometer and intelligent wearing equipment) to collect data of blood pressure, blood sugar, heart rate, exercise, sleep and the like, the health information investigation is mainly based on a paper questionnaire or an electronic questionnaire, and relates to dietary structure of an individual, drug use condition, past medical history, psychological state and other behavior related information, and part of early warning systems try to mediate the two types of data with a preset health assessment model and output preliminary health risk prompt. The prior individual health early warning technology has a plurality of defects in practical application, and is characterized in that firstly, obvious fragility exists between physiological monitoring data and behavior questionnaire data, the prior art is used for independently collecting two types of data to establish an effective data fusion mechanism, so that the health assessment dimension is single and a complete individual health portrait cannot be formed, secondly, the prior physiological monitoring equipment has a complex operation flow aiming at a special group (especially the elderly), the behavior questionnaire is mainly made of professional medical terms and depends on individuals to fill in independently, the aged generally has the problems of weak operation capability, poor memory, difficult understanding of the professional terms and the like, and misoperation, filling omission, filling misplacement and the like of data are easily caused, so that the defects of data accuracy and continuity are caused. Therefore, an early warning method and system capable of realizing deep fusion of physiological monitoring and behavior questionnaire data and adapting to special group requirements are required to be developed. Disclosure of Invention The invention aims to solve the technical problem that misoperation, missing filling and incorrect filling of data acquisition are easy to occur in the process of individual health early warning for the elderly, and provides an individual health early warning method and system for fusing physiological monitoring and behavior questionnaires. The technical scheme adopted for solving the technical problems is as follows: On the one hand An individual health early warning method integrating physiological monitoring and behavior questionnaires comprises the following steps: S1, personalized configuration and multi-source data acquisition, namely, based on a user individual file, self-adapting a data acquisition mode, acquiring physiological time sequence data through household monitoring equipment, and acquiring behavior data and environment data through a dynamic interactive questionnaire; S2, data fusion and personal health knowledge graph construction, namely preprocessing and semantically analyzing the acquired physiological time sequence data and behavior data, and constructing and dynamically updating the personal health knowledge graph based on the preprocessed physiological time sequence data, behavior data, environment data and a medical knowledge base, wherein the personal health knowledge graph takes a user entity as a core, associates physiological index nodes, behavior event nodes and environment factor nodes, and defines time sequence, cause and effect and statistical association relations among the nodes; S3, risk assessment based on digital twinning and personal health knowledge graph, namely establishing a personal health digital twinning model of a user based on the personal health knowledge graph, carrying out multi-dimensional and hierarchical dynamic health risk assessment on the user by utilizing the personal health digital twinning model and a risk assessment algorithm, and outputting a risk assessment result and a risk tracing path; and S4, personalized intervention and closed loop verification, namely generating and pushing a personalized health intervention plan according to the risk a